Lies, damned lies and statistics

On 13th January 2014, the Ministry of Justice was sent a report produced by Oxford Economics, an independent consultancy which had been jointly commissioned by the Law Society, the London Criminal Courts Solicitors Association, The Criminal Law Solicitors Association, the Legal Aid Practitioners’ Group and the Big Firms’ Group. Oxford Economics was commissioned to develop a clearer understanding of how the Legal Aid Agency (LAA) forecasts criminal Legal Aid expenditure and to produce an alternative forecast based on plausible assumptions of future trends in crime and the criminal justice system. It was commissioned based on comments made by Chris Grayling at a meeting with various parties before Christmas after which it was felt that some independent research was necessary.

The report shows that at the police station, Magistrates’ Court and Crown Court the number of cases being handled under legal aid is falling. Despite the obvious fall in numbers, the Government uses a flat line on which to base its estimates of future levels of cases coming through the system. The report suggests that by 2018/19, based on the falling number of cases coming into the system, the Government would save about £84 million by doing absolutely nothing. This is a relatively simplistic précis of the report but sets out the main headlines.

The Lord Chancellor wrote to some of the organisations who commissioned the report on Friday rejecting it outright. This may be no more than one might expect from a man who simply refuses to listen to reasoned argument but the manner of his rejection demonstrates a cavalier attitude to the use of figures and statistics. You may recall that he was censured in 2010 for using statistics in a way which was likely to mislead the public (see http://news.bbc.co.uk/1/hi/uk_politics/8498095.stm). The method of recording violent crime had changed in 2002, making the figures for the periods before and after that date non-comparable and Grayling’s quotes led Sir Michael Scholar, chairman of the UK Statistics Authority, to write to Mr Grayling saying: “I do not wish to become involved in political controversy, but I must take issue with what you said yesterday about violent crime statistics, which seems to me likely to damage public trust in official statistics.” In notes attached to the letter, the statistics authority said it regarded “a comparison, without qualification, of police-recorded statistics between the late 1990s and 2008/09 as likely to mislead the public”.

In his response to the Oxford Economics report Mr Grayling makes two comments of dubious value:-

Firstly, he says, “Given that our own forecast last year was accurate to within 1%, there would need to be a compelling reason to change our forecasting methodology for the future.”

From this comment one can only surmise that Mr Grayling is claiming that the LSC (as it then was) forecast for 2012/13 was accurate to within 1% for the criminal spend on legal aid (the report only deals with crime spend). The forecast for the crime spend under legal aid made by the LSC was £1.025 billion according to the LSC Business Plan for 20-12/13 (http://www.justice.gov.uk/downloads/publications/corporate-reports/lsc/lsc-business-plan-12-13.pdf). The actual spend, according to the LSC Annual report (http://www.justice.gov.uk/downloads/publications/corporate-reports/lsc/lsc-annual-report-12-13.pdf), was £975 million. The Government was therefore £50 million out on their forecast. In order to calculate this as a percentage, one undertakes the following calculation:-

50,000,000/1,025,000,000 x 100 = 4.88%

By this basic calculation, the forecast of the Government was inaccurate to the degree of 4.88% or in round terms 5 times what the Lord Chancellor claims to have been the case in his rejection of the Oxford Economics report.

The second comment which may merit some closer examination is: “Since 1997, incidents of crime have decreased by 50%, whilst criminal legal aid spending has increased by around 10% in real terms.” Mr Grayling should be aware after his brush with the UK Statistics Authority that there is an inherent danger in seeking to use comparisons of crime figures from periods before and after 2002.

However, this is perhaps not the main issue. The level of “incidents of crime” is taken to mean the rate of reported crime. It is not the sort of expression that would be used if he were talking about “acts of assistance” under criminal legal.

To anyone who understands the criminal justice system, the level of reported crime is largely irrelevant. What counts in terms of spending on legal aid is the number of cases coming into the system whether it be at the police station or into the courts. The level of reported crime has very little correlation it seems with the level of cases coming into the system. The Oxford report shows that over the last few years these levels are falling. This may be part of the reason why the amount spent on criminal legal aid has fallen each year for the last decade. It is, of course, not the only reason as the changing structure of payments, always to the detriment of the lawyers representing defendants, and the continual cuts in rates and fees has much to do with it as well.

It is difficult to find reliable figures to argue against the claim that legal aid spend has risen 10% in real terms since 1997. In fact, it appears that this may be correct. However, what is undeniably correct is that for the last 10 years the spend has fallen every year. There may have been an increase in costs between 1997 and 2003 but that rise is likely to be explainable by the myriad changes brought in by the then Government, creating over 3000 new offences and introducing measures which increased the work required on every case (e.g. admission of bad character and hearsay). At that stage, solicitors were paid on hourly rates and not fixed fees at the police station and in the Crown Court and were paid full fees for committal hearings so extra work on cases meant extra cost. At the same time a number of matters have since been taken out of scope such as general advice and assistance under the old Green Form/Claim 10/CDS1 and 2 scheme. Given the fundamental changes in payment structures (like the changes for the reporting system for crime rates), a like for like comparison of costs is not necessarily instructive or accurate.

However, 1997 does have some relevance for criminal legal aid lawyers as it is the last time that any of us can remember getting an increase in rates of pay. The “direction of travel” for legal aid spend in crime has been relentlessly downhill for a decade and that should be considered statistically significant. Like the Oxford Economics report, it is not something that the Lord Chancellor should be ignoring.

Leave a Reply

Your email address will not be published. Required fields are marked *